Each year, millions of Americans are admitted to an intensive care unit (ICU), which can initiate a cascade of decisions about treatment and end of life care. This is particularly the case for patients with chronic critical illness. The chronicall critically ill (CCI) are at high risk for cognitive impairment, prolonged use of life- sustaining cre, and the need for a surrogate decision maker (SDM) to make decisions about end of life. SDMs of the CCI often describe high states of psychological stress associated with the uncertainty of the patient's condition and their decision making role. Many SDMs find it difficult to comprehend, process, and apply information from clinicians to make informed decisions for their loved ones. Presently, SDMs have no effective tools to help them make informed decisions about the care of a cognitively impaired CCI patient at end of life. Thus, we propose to revise and test the efficacy of an avatar-based decision support technology referred to as Interactive Virtual Decision Support for End of Life and Palliative Care (INVOLVE). This project has two phases. In Phase 1, we will revise our INVOLVE prototype through an iterative, user-centered process. Phase 2 is a three-arm, clinical trial of 270 SDMs of cognitively impaired CCI patients, comparing INVOLVE versus information-only support (IS) and usual care (UC). The specific aims for this project are: [1] identify the essential elements of the graphical user interface and educational content needed to revise the INVOLVE prototype for a set of common end of life decisions that occur in the ICU; [2] evaluate if there are differences in the decision making readiness and decision making quality between subjects exposed to INVOLVE, IS, or UC on Days 1, 3, and 7 days post-baseline while accounting for covariates (prior SDM experience, SDM knowledge of the patient's preferences, and SDM's religious beliefs); and [3] determine if there are differences in the post-decision outcomes of SDMs and their CCI patient by study condition while accounting for covariates at 90 days post-baseline. The importance of finding strategies to improve the quality of decision making for end of life care in the ICU has been well recognized. This will be the first study to test interventions tailored to the unique needs of the SDMs of CCI patients delivered using an interactive avatar based format. This is a format that is easily delivered, well accepted, and well suited for this vulnerable and difficult to reach cohort f decision makers. If proven efficacious, these technology-based interventions for end of life decision making may also offer sustainable solutions that can be easily adapted to varied populations to improve the outcomes of SDMs and their cognitively impaired loved ones.